Applying a re-emergence model to inform weed monitoring visitation rates

摘要

To eradicate an invasive plant species from a site it is necessary to destroy all individuals and monitor the site to prevent re-establishment from the soil seed bank or other populations. Quantifying re-establishment potential after control works canhelp to allocate monitoring resources and determine when to declare a monitored site eradicated. We propose using a recurrent event analysis to model the potential for individual re-emergence at controlled sites. This approach is applied to the monitoring program for controlled hawkweed (Pilosellct spp.) infestation sites at Falls Creek, Victoria. We use the re-emergence model to guide decisions on reducing monitoring frequency for sites with low re-emergence probability. The decision criteria to reduce visitation of sites are presented as simple, conservative rules of thumb. We simulate these stopping rules using historical observation data to assess the accuracy of decisions. We found that 52% of all sites being monitored in the 2012/2013 season satisfied the reduced visitation criteria, which could have produced a resource saving of 154 person days if the monitoring frequency was reduced from once per week to once per month. Simulations show that re-emergence can occur outside the predictive capabilities of the model, although the accuracy of decisions improves with longer monitoring periods.